Permian depositional age of metaturbidites of the Duque de York Complex, southern Chile: U-Pb SHRIMP data and palynology

The Duque de York Complex (DYC) is par! oflhe low gmde metammphic accretiooary complexes oflhe pre-Andean Patagonian 'basement'. It is a sedimentary succession exposed along the western margin of southemmost SoulhAmerica. New U-Pb zircon ages aod palynological data restrict!he maximum depositiooal age oflhe DYC to Ihe limit betweeo Ihe early Permian (Kunguriao) aod Ihe middle Permian (Roadiao). The palynological association recorded in Ihe DYC, cbaracterized mainly by Gymnospennopsida pollen, indicates a humid environment offorest wilh an undergrowth offems. Regional paleogeograpbic corrclations point out that 3D intcrprctation ofDYC as an autochthonous terranc caonot be discarded, cootrasting wilh previous hypolheses which snggest ao allochlhooous character for !bis complex.


Introduction
The Duque de York Complex (OYC) is one ofthe metamorphic complexes that fonn the pre-Andean Patagonian 'basement' rocks, which crop out extensively along the westem edge of South America south of 50 0 S.These rocks have been generally considered as par! of the late Paleozoic-early Mesozoic accretionary prism built at the paleo-Pacilic (panthalassan) margin ofGondwana (e.g., Hervé el al., 1981;Forsythe, 1982).The accretionaryorogenic belt lhat fonned on tbis margin is one of the largest known orogenic belts in Earth history, and now occupies the eastem third of Australia, New Zealand, West Antarctica, the Transantarctic Mountains and large parts of southem South America (Vaughan el al., 2005).This orogenic belt has been tenned in two different ways depending on the interval considered: the Proterozoic and Paleozoic TerraAustralis orogen (Cawood, 2005), and the Paleozoic and Mesozoic Australides (Vaughan el al., 2005).It has been regarded as a collage of accreted !erranes-!erranes being fault-bounded packages of rocks of regional extent characterized by a geological history that differs from lhat of neighboring !erranes (Howell el al., 1985;Vaughan el al., 2005).
FOT a long time, the accretionary complexes that compose the MDAC have been considered of exotic or allochthonous origin, or at least, as suspect terrones.Terranes are 'suspect' if there is doubt about their paleogeographical selting with respect to adjacent terranes or continental margin (Coney el al., 1980;Coombs, 1997), and may be described as 'exotic', 'far-travelled' or 'alIochthonous' (all meaning about the same thing) if there is sufficient evidence that they originated far from their present locations, often assumed to be hundreds OT thousands ofkilometers away (Vaughan el al., 2005).The consideration of the MDAC as suspect and potentially exotic is based mainly on its fossil content and the inferred depositional selting for thero (in the case of the TL and the coeval DC) (e.g., Ling el al., 1985;Ramos, 1988) and on the impossibility to fmd a conteroporaneous magmatic are as the souree of Pennian zircons for the DYC anywhere at a similar latitude in southem Patagouia (Hervé el al., 2003;Hervé and Mpodozis, 2005;Hervé el al., 2006).The latter has led to propose lhat the deposition ofthe DYC took place at high southem latitudes along the Antarctic sector of the Gondwana margin (Lacassie, 2003;Lacassie el al., 2006).However, derivation of the DYC from lower and warmer latitudes is a hypothesis that canoot be ruled-out.
The lack ofindex fossils in the DYC has prevented an accurate determination ofthe depositional age of tbis complex.In consequence, a late early Pennian maximum depositional age has been established by the use ofthe youngest detrital zircon population in these metasediments (Hervé el al., 2003).Nevertheless, the use of the youngest detrital zITcon age or population in a sediment as a limit for the age of deposition has been questionable for both geological (there is no necessary connection between the timing ofzircon-generating events in a souree region and the age of fina1 deposition of a sediment eroded from tbis souree) and statistical reasons (Andersen, 2005).In tbis context, tbis paper presents the frrst palyoological stody in rocks ofthe DYC and it also corresponds to the liTst record of late Paleozoic palyoomorphs in Chile.The aim of tbis work is to restrict the age range ofthe DYC by the combination ofthe palyoological results with new U-Pb SHRlMP ages in key samples, and also gives a revision of the paleoenvironmental and geochronological data regarding the place and timing of deposition of the DYC.The conclusions derived from tbis work allow giving some new considerations focused on the supposed alIochthonous character ofthe MDAC, and especially of the DYC.
U-Pb SHRIMP detrital zircon ages from sandstones ofthe DYC reveal lhat lhe youngest main opulation, and hence lhe maximum possible depositional age, is late early Permian (ca.270 Ma) (Hervé et al., 2003).The geochemical study of Lacassie et al. (2006), complementing lhe data and refining lhe conclusions ofFaúndez et al. (2002), indicates lhat lhe DYC sandstones and mudstones had lheir source in a volcanic are of granodioritic average composition located relatively proximal to the depositional basin, and whose plutonic roots had been exposed by erosiono AIso, lhey propose lhat lhe DYC was deposited in a tectonic setting corresponding to an active continental margin, possibly located along lhe Antarctic segment of lhe Panlhalassan Gondwana margino
The apparent lack of a Permian magmatic arc in soulhernmost Patagonia allowed Lacassie (2003) and Lacassie el al. (2006), following Hervé el al. (2000) and Cawood el al. (2002), to propose lhat lhe accretion of lhe TL and lhe DC would have occurred against lhe Antarctic-Australian segment of lhe Gondwana margin, from where bolh would have been displaced by dextral translation, togelher wilh lhe DYC, as a coherent block to lheir current position.In addition, Lacassie el al. (2006) show that the DYC metasediments share important petrographic, geochemical and geochronological characteristics wilh metaturbidites present in lhe Rakaia Terrane in New Zealand and wilh lhe eastern (Triassic) Le May Group in Alexander Island.These similarlties point tuwards similar igneous sources for lhe three successions, suggesting lhat lhey were coevally deposited along lhe same active continental margin (Lacassie el al., 2006).This margin was probably located along lhe Antarctic sector of lhe Panlhalassan Gondwana margin, as favored by lhe studies of Willan (2003) for the source area oflhe Le May Group, and ofWandres el al. (2004) and Wandres and Bradshaw (2005) for lhe source area of lhe Rakaia terrane.The last two sludies indicate !hat lhe origin of lhe Permian detritus in lhe Rakaia terrane would be in lhe igneous rocks of lhe Amundsen and Ross Provinces, East Antarctica, which during lhe Permian were close to 60 0 S (Veevers, 2004;Cawood and Buchan, 2007).If lhe source of Permian detritus was lhe same for lhese three successions (DYC, Rakaia Terrane and Le May Group), this would imply dextral strikeslip displacement of lhe MDAC along lhe SW Gondwana margin from lhese high latitude to its present position.
The deposition of sediments of lhe DYC in high soulhem latitudes (Lacassie, 2003;Lacassie el al., 2006), contrasts wilh the second scenarlo, which involves deposition of lhe DYC in lower and warmer latitudes, perhaps associated with subsequent sinistral strike-slip movements of lhe entire MDAC along the Panlhalassan margin of Gondwana.

Sampling and Methods
Palynological data were acquired from one sample oflimestone of lhe TL and seven samples ofmetasediments oflhe DYC, five from Madre de Dios Archipelago and two from Diego de Almagro Archipelago (Table 1; Fig. 1).All samples were processed by standard palynological melhods.All but one of lhe samples (MD05-20 from Guarello Island) yielded poorly preserved palynomorphs.The study and lhe description of lhe specimens were made wilh an optical microscope.The slides are housed at lhe Laboratory ofPaleopalyoology oflhe Departamento de Ciencias de la Tierra, Universidad de Concepción under codes 1396 to 1401.
Two metasedimentary samples (F004-21 and F004-22) from lhe units oflhe MDAC were collected for U -Pb zircon dating by SHRIMP RG (sensitive high resolution ion microprobe, reverse geometry) at lhe Research School ofEarth Sciences, The Australian National University.Zircon grains were separated from total rock samples using standard crushing, washing, heavy liquid and paramagnetic procedures.The zircon-rich heavy mineral concentrates were poured onto double-sided tape, mounted in epoxy togelher wilh chips oflhe reference zircons (FCI and SL13), sectioned approximately in half, and polished.Reflected and transmitted light photomicrographs were prepared for all zircons.Calhodoluminescence (CL) Scanning Electron Microscope (SEM) images were prepared for all zircon grains.The CL images were used to decipher the intemal s!ructures of lhe sectioned grains and to ensure lhat lhe -20 11m SHRlMP spot was wholly wilhin a single age component (usually the youngest) wilhin lhe sectioned grains.
The U-Th-Pb analyses were made using SHRlMP RG.The zircon grains were analyzed sequentially and randomly.Each analysis consisted of 4 scans through lhe mass range, wilh a reference zircon analyzed for every five unknown zircon analyses; SHRlMP analytical method follows Williams (1998, and references therein).The data have been reduced using the SQUID Excel Macro ofLudwig (2001).The U-Pb ratios have been normalized relative to a value ofO.01859 for the FC I reference zircon, equivalent to an age of 1,099 Ma (Paces and Miller, 1993).Uncertainties given for individual analyses (ratios and ages) are at the one sigma level (Tables 2 and 3).Tera-Wasserburg concordia plots, probability density plots wilh stacked histograms and weighted mean 206Pbl'''U age calculations were carried out using ISOPLOTIEX (Ludwig, 2003).The 'Mixture Modelling' algorilhm of Sambridge and Compston (1994), via ISOPLOTIEX, was used to un-mix statistical age populations or groupings; from lhese groups weighted mean 206Pbl'''U ages were calculated and the uncertainties are reported as 95% confidence limits.
An estimate for the maximum age for the deposition of the sediment sample may be determined from the weighted mean age of the youngest peak in these distributions, where ~3 analyses are wilhin analytical uncertainty.Such an age grouping has taken into account isolated cases of inferred radiogenic Pb-Ioss, which can produce minor scatter to younger ages.Ages for individual grains are reported at lhe 68% confidence level, and Geological Time Scale referred throughout lhe text is lhat of Gradstein el al. (2004).

Palynology
The palynological anaIysis revealed a palynoflora composed predominantIy by Gymnospermopsida monosaccate pollen grains, altbough Gymnospermopsida bisaccate pollen grains were also observed.Selected species are illustrated in figure 3. The samples show a very low frequency of palynomorphs, and in most cases an exact identification of tbe species is impossible because of tbe bad preservation state of tbe palynomorphs.The palynomorphs detected within tbe TL are Pune/atispori/es pune/atus (Ibrahim) Ibrahim, a Pteridophyta known from tbe Carboniferous lo Triassic in New Zealand, Australia, Asia, Europe and Soutb Amerlca (Alpem and Doubinger,  1973; Owens e/ a/., 2002; Pérez Loinaze, 2008).

Petrography
The metasedimentary samples collected for U-Pb zircon dating were obtained from an outcrop where the OC and the OYC are in conformable and intercalated stratigraphic contact (Fig. 4).Significantly, this site corresponds to one of only two localities were this type of contact between these complexes is recorded.The samples were spatially associated, stratigraphically separated by ca. 10 m.The fust sample (F004-21) comes from a deformed (folded) metasedimentary horizon (0.04 to 0.06 m thick) of toffaceous character, interbedded in metacherts of the OC.The microscopic petrographic description of this sample shows that it is mainly composed by very angular fragments (0.01-0.2 mm) of quartz (55%), altered feldspars (30%) and biotite flakes (15%) in a cryptocrystalline siliceous matrix.Accessory minerals include zircon, game!, sphene, apatite and Fe-oxides.Scarce small and highly altered shard fragments were also observed.The biotites are Al oriented parallel to the contacts with the underlying radiolarían chert.This last featore coupled with the normal grading observed in this bed agrees with subaquatic conditions of deposition.The second sample (F004-22) is a quartz rich metasandstone of the OYC prevíously analyzed by palynological methods.The sample was extracted from a massive and structureless sandstone bed (20 m of minimum thickness) nearly 3 m aboye the contact with the banded cherts of the OC.Jt is a feldspathic arenite formed by well to moderately sorted subangnlar and highly spherical fragments, with sizes between 0.02 and 1.2 mm (0.3 mm in average).Main fragments are quartz (60%), feldspars (30%), biotite (8%) and white mica (2%).Accessories include zircon, apatite, lithic fragments (basalts and rhyolites), gamet, pyrite and Fe-oxides.

U-Pb SHRIMP ages of detrital zircons
The Tera-Wasserburg diagrams plot the total ratios, uncorrected for common Pb, and show that the data generally plot close to Concordia (Fig. 5).Relative probability spectra of the detrital zircon ages are presented in figare 5.For sample F004-21 54 grains were analyzed, whereas 42 grains were exarnioed for sample F004-22.F004-21.The zircons ofthis sample are prismatic and euhedral crystals, wilh zoned magmatic interna!structures as seen under CL imaging (Fig. 6).This is compatible wilh its textural and mineralogical characterlstics, which are indicative ofthe tuffaceous character of lhe metasediment.AIlhough some of lhe youngest individual ages involve significant common Pb correction (Fig. 5); a correction has been applied to derive lhe radiogenic ratios and age oflhese analyses (rabie 2).The age spectnun shows a narrow range of provenance ages wilh a major peak in lhe early middle Perntian, representing ca.---------------  There are scattered older ages ranging from Early Paleozoic to Neo and Mesoproterozoic aged noise and one Paleoproterozoic aged zircon.A weighted mean 206pbl"'U age of289.7±2.1 Ma (MSWD= 1.3) place a constraint on lhe maximum age of deposition of lhis metasediment.
Palynomorphs documented in tbis study are comparable lo tbose recorded in Carboniferous and Permian strata of other late Paleozoic Gondwanan basins, such as those exposed in tbe Chaco-Panmá Basin in Argentina and Uruguay (Césari el al., 1995;Beri el al., 2006), and at Rio Grande do Su! in Panmá Basin, Brazil (Souza and Marques-Toigo, 2005).It is noteworthy lo ernphasize !hat tbe only palynomorph with exclusive Permian record (Praecolpatiles sinuosus) is in one of tbe samples witb a Permian maximum depositional age (Fig. 4).Regarding the paleoenvironrnental background mentioned above and the Permian age obtained in tbe DYC, tbe proposed paleoc1imatic selting for the deposition ofthe late Paleozoic basins of soutbwestem Gondwana, in particn1ar tbose situated in Patagonia, helps to establish regional correlations and sorne paleoc1imatic inferences.Archangelsky el al. (1996) and Limarino el al. (1996) suggested a specially humid and temperate climate, even subtropical, for the deposition oftbe La Golondrina Formation (the lower mernber in tbe La Golondrina Basin, Patagonia), whose age wou!d be restricted between the Sakmarian and the Kungurian (Limarino and Spalletti, 2006).The age and paleoclimatic features of tbis formation are remarkable, particu!arly because it crops out in a relatively clase position (ca. 500km apart) to tbe present-day position of tbe outcrops of tbe DYC (Fig. 7).Another late Paleozoic Patagonian basin that shares similar paleoclimatic characteristics and also an adjacent location lo the deposits oftbe DYC is the Tepuel-Genoa Basin (López-Gamundi andLimarino, 1984; Andreis el al., 1987) (Fig. 7), likely related to metamorphic rocks outcropping in tbe Coastal Range of Chile (Hervé, 1988;Duhart el al., 2001).The early Permian component of this basin (Sakmarian-Artinskian, according to Césari el al., 2007) is represented by the Río Genoa Formation, whose sediments have also been interpreted as deposited in a humid and subtropical climate (Archangelsky el al., 1996; Limarino el al.,   1996).However, it is not possible to determine a more precise paleoclimatic connection with the La Golondrina Formation, mostly because late early Permian (Kungurian) sedimentary rocks have not been identified in the Tepuel-Genoa Basin (Limarino and Spalletti, 2006).If it is assumed a fixed position ofthe DYC with respect 10 Patagonia since its deposition, and considering the previous exam-pIes and the age obtained for the DYC, it would be probable that warm paleoclimatic conditions were recorded in the metasediments of this complexo This was, however, impossible to register in this s1ody, mainly due to the very low proportion of palyoological material in the samples.Even though we canoot precise the paleoclimatic conditions during the deposition ofthe DYC, sorne observations can be done.The types of deposits accumulated as a consequence of ice activity are very varied, and severa!noo-glacially related mechanisms can produce similar deposits.The recognition of these accumulations is more difficult if we consider that glacial deposits are frequently reworked in outwashes or by mass flow.For these reasons, the assigument of a glacial origin 10 any deposit, or its refutatioo, needs the combinatioo ofthe properties of the deposit itself, the adjacentrocks units, as well as climatic and paleogeographic conditions at the time of depositioo (e.g., Charrier, 1986).Additionally, 10 establish that a sedimentary succession, or part of it, preserves a record of glacial, proglacial or periglacial depositional environments, multiple facies criteria are needed (diamictites, chaotic fabrics, rhy1lnnites, laminated mudrock:s witb outsized dispersed clasts (lonestones), striated pavement, and faceted, hulletshaped and striated clasts, among otbers; Miller, 1996), or it is required a single criterion !hat unambiguously appears to indicate glacial influence by virtue of its occurrence out of context witb enclosing facies (e.g., Fielding el al., 2008).In this context, even tbough tbere have been records of diamictites in tbeDYC (Cecioni, 1956;Forsythe andMpodozis, 1983), this does not give conclusive evidence of a glacial environment of deposition.Moreover, tbere are ouly few localities in tbe whole extension ofthis accretionary complex (more tban 1,000 km') where diamictites have been identified, but tbe existence of glacial characteristics sensu striclo like lonestones (dropstones), striated pavements or faceted clasts have never been reported.
The large extension and volume of tbe DYC could be attributed to tbe great volume of fresh water produced by ice-melting during tbe waning of tbe glaciation in Gondwaoa (Buatois el al., 2006).
This huge volume ofwater probably reworked large amounts of sediments !hat were subsequently deposited at tbe margins oftbe continent.The ichnofaunas registered in DYC (Scalarituba isp., Chondriles isp., Planoliles isp., Palaeophycus isp.andAncorichnus isp.; in Lacassie, 2003), are distinctive of marine environments (written communication, L. Buatois, January 2008), and tbe low content of palynological material in tbe samples suggests a marine offshore environment of deposition of tbe host rock:s.Botb proposals match witb tbe interpretation oftbe DYC rock:s as turbidites, and would exclude a fjord-like setting for tbe erigin oftbis succession and indicate !hat tbe place of deposition of tbe DYC was located far away from tbe direct influence of fresh water formed during tbe Gondwanan deglaciation.
The data obtained for sample F004-22 of tbe DYC (289.7±2.1 Ma) indicate an early Permian (Sakmarian) maximurn possible depositional age.This is nearly 20 m.y.older!han tbe youngest U-Pb SHRIMP ages compooent recorded for detrital zircons in tbe DYC (Hervé el al., 2003).Nonetbeless, tbe former autbors have recognized this Sakmarian peak (ca.290 Ma) as an individual Permian population in tbe age spectrum of tbe detrital zircons in tbe DYC.The apparent inverted stratigraphical position of tbe analyzed samples (Fig. 4) can be explained as F004-21 being tbe airbome volcanic material deposited near tbe continental margio ofGondwana, and F004-22 as resedimented detritus, formerly deposited somewhere between its local source area and tbe final depositional site, and tben redeposited as turbidite flows aboye tbe cherts which include tbe tufIaceous layer represented by F004-21.44°S, approximately) can be identified as tbe most important souree so far recognized for tbe provenance of detritus in tbe late Paleozoic metasedimentary rock:s along tbe Pacific margio of Gondwana.This would imply transport by a system oflong rivers over a wide and relatively flat pre-Andean platform (Hervé el al., 2003), or even eolian transport over hundreds of kilometers (e.g., Dickinson and Gehrels, 2009).
The widespread abundance of radiometric data frem ash fall deposits within tbe range of 280 to 260 Ma has been conunouly attributed to a period of intense silicic volcanism along tbe continental margin of soutbwestem Gondwaoa that peaked around tite 270 Ma (López-Gamundi, 2006).Besides tite model of Pankhurst el al. (2006), until now no analogous magmatic process responsible for tite ubiquitous Sakmarian peak (ca.290 Ma) in tite U-Pb spectrum from samples oftlte OYC has been clearly identified.Similar and equivalent radiometric ages have been obtained from adjacent late Paleozoic basins of soutltwestern Gondwana.In tite soutltemmost Karoo Basin of Soutlt Africa, Bangert el al. (1999)  Gondwana around the Carboniferous-Permian boundary (ca.299 Ma).It is, however, difficult to restrict tite magmatic activity in soutltwestem Gondwana lo a single instant in Permian times, since most of tite limited available data are in tite range of 270 lo 290 Ma and usually tltey overlap witltin tlteir analytical uncertainty.Instead, tite entire early Permian could be regarded as a period of active and geographically widespread magmatism in this region of Gondwana.This scenario would also explain tite strong coincidenee oftlte main Permian peaks in tite detrital zircon U -Pb spectrum of tite metasediments of tite OYC, tite Rakaia Terrane (New Zealand) and tite eastern Le May Group (Alexander rsland, Antarctica) (Lacassie el al., 2006).Nevertheless, tite cause of the petrographic and geochemical similarities between tltese three metasedimentary uuits is an issue not discussed here as it is beyond tite scope of tltis work.

Paleogeographic correlations
So far, tltcre has been no consensus in tite place of accretion of tite TL and OC and tite supposed subsequent sense ofmovement oftlte MDAC (as a coherent block) along tite soutltwestern Gondwana margino Lacassie el al. (2006) proposed accretion of tite OC-TL assemblage at tite Antarctic-Australian portion of the Gondwana margin, followed by dextral translation of tite MOAC (and hence of tite OYe) parallel lo tite margino On tite otlter hand, tite possibility of a virtually fixed position for tite OYC sinee its deposition (witlt Patagouia as reference) is hampered by tite fact tItat tltere is no clear indication of a coeval Permian magmatic are at latitode similar lo tite current position of tite MDAC in soutltem Patagouia (e.g., Hervé el al., 2006).This magmatic are could be represented eitlter by tite Choiyoi acid magmatic provinee (Kay el al., 1989;Mpodozis and Kay, 1990) or by tite Permian igneous rocks in tite North Patagouian Massif (pankhurst el al., 2006), botlt sources currently located north of tite 40• and 44•S, respectively.However, tite Choiyoi Formation presents ages of ca.281 Ma near its base (Rocha-Campos el al., 2006;Suárez el al., 2009), and tltus cannot account for tite 290 Ma peak in tite U-Pb age spectrum.Conversely, it is well known tltat tite subduction ofbatltymetrically elevated oceanic featores such as ridges or plateau (DC and TL in this case) can flatten tite subducting slab and prevent tite magmatic activity in tite vicinity of tite continental margino According lo tite above-mentioned situation, tite OYC would not have had displacement since its deposition and tite remnants of tite associated Permian magmatic are could still be hidden below tite Mesozoic sedimentary cover somewhcre in tite soutlteastern Patagonia or even farther eastward.This option has been recently explored by Ramos (2008), who proposes a late Paleozoic magmatic are witlt a soutltem extension in tite NNE trending Río Chico-Punta Dúngenes High (Fig. 7).This hypotltesis would preelude defining tite OYC as an allochtltonous terrane.A precedent tItat supports this hypotltesis is tite presenee of metasedimentary rocks witlt detrital zircons witlt U-Pb SHRIMP ages similar (ca.290 Ma) lo tite ones recorded in tite OYC lo tite east oftlte Soutlt Patagonisn Batltolith, at almost tite same latitode oftlteMDAC (Augnstssonel al., 2006).In brief, additional and more detailed paleomagnetic, geochronological and isotopic work is needed lo give more convincing arguments supporting one of tite proposed hypotlteses, an autochtltonous or an alIochtltonous origin oftlte OYC.
Little is known about tite exact paleogeographic confignration of tite Antarctic Peninsula during tite late Paleozoic, tltough one of tite most accepted current paleogeographic reconstructions locate tite Antarctic Peninsula lying west of Patagouia in tite Middie Jurassic (Kiiuig and Joka!, 2006).Thcrefore, tite possibility of a geological correlation between tite geologic uuits present in westem Patagouia and tltose exposed in tite Antarctic Peuinsula must be considcred.In tltis contex!, it has been suggested based on tite similarity oflitltology and detrital zircon age pattems, 1ba1 the Trinity Peninsula Group (fPG) is the equivalent COWlterpart ofthe DYC in theAntarctic Peninsula (Hervé el al., 2006) (Fig. 7).Willan (2003) assumed that the TPG could have been derived from a glaciated continental margin, though bis result is hased ouIy on indirect evidence (geochemical weathering) and, at thls time, there are no palynological reports on this unit According lo the paleogeographic recons1ructions fur the late Paleozoic (e.g., Cawood andBucban, 2007), the TPG was supposedly located in a higher paleolatitudinal position than the late Paleozoic Patagouian hasins, and therefore the paleoclimatic conditions wouldhave been colder during its deposition.However, caution must be taken with thls interpretation, mainly because the TPG is part ofthe Westem Domain ofthe Antarctic Peninsula (Vaughao and Storey, 2000), which has been regarded as a suspect terrane and even as allochthonous lo the rest ofthe !erranes ofthe Antarctic Peninsula (Willan, 2003).

Concluslons
This contribution presents the first palynological record fOl the late Paleozoic in Chile.The palynological assemblage recOlded in the DYC is composed mainly ofGynmospermopsida pollen, with also Pteridophyta and fungal spores.The studied association indicates a humid environment of forests with an Wldergrowth offerns.
The palynological data indicate a Permian age for the deposition ofthe DYC.This age is also supported by new U-Pb SHRIMP detrital zircon ages, which constrain the maximum depositional age of the DYC lo the limit between the early Permian and the middle Permian (ca.270 Ma), confirming the maximum depositional age obtained by previous geochronological data (Hervé el al., 2003).
The available data indicate that the allochthonous hypothesis for the DYC is not completely proved, and an autochthonous tectonic setting (with respect to Patagonia) could also be a possible interpretation.
FIG. 5. Tcra-Wasscburg diagrams for zircon U-Pb data.AnaIyscs are plotted as total ratios calibratcd for U-Pb, but uncorreeted ror common Pb.The error ellipses are 68.3%confidence limits.Th.e dotted arrow in F004-21 shows the diroction of common Pb at 270 Ma.Alignment along this arrow suggests a real inferred age on Concordia with variable degrecs of incorporated common Pb at the time of crysta11ization.
The grains analyzed from this sample are very low in common Pb.The relative probability plots oflhe detrital zircon ages display a prominent component in lhe early Perntian (ca.40% of the analyses), wilh olher subordinate peaks in lhe Carboniferous and Devonian (ca.17% of lhe analyses each one).

an active and extensive volcanic event in westem
reported 288.0±3.0 and 289.6±3.8Ma from bentouitic tuffbeds intercalated in sedimentary rocks oftlte Prince Albert Formation, lower Beca Group.Rocha-Campos el al. (2006) and Goerra-Sommer el al. (2008) acquired comparable Permian ages from ash-fall deposits interbedded in coal successions of tite soutltem Paraná Basin in Brazil.The last autltors claUn tltat this information supports tite presence of